航空学报 > 2007, Vol. 28 Issue (3): 673-677

基于径向基神经网络干扰观测器的空天飞行器自适应轨迹线性化控制

朱亮,姜长生,张春雨   

  1. 南京航空航天大学 智能控制实验室
  • 收稿日期:2006-05-22 修回日期:2006-11-06 出版日期:2007-05-10 发布日期:2007-05-10
  • 通讯作者: 朱亮

Adaptive Trajectory Linearization Control for Aerospace Vehicle Based on RBFNN Disturbance Observer

ZHU Liang,JIANG Chang-sheng,ZHANG Chun-yu   

  1. Intelligent Control Laboratory, Nanjing University of Aeronautics and Astronautics
  • Received:2006-05-22 Revised:2006-11-06 Online:2007-05-10 Published:2007-05-10
  • Contact: ZHU Liang

摘要:

研究了一种自适应轨迹线性化控制策略并应用于空天飞行器(ASV)飞行控制系统设计。通过理论分析指明当前轨迹线性化控制方法(TLC)对系统中的不确定存在鲁棒性不足的问题。为了解决这一问题,首先研究了一种径向基神经网络干扰观测器(RDO)技术,严格证明了RDO对于系统中不确定因素具有良好的逼近能力。然后利用RDO输出得到一种新的基于RDO的自适应TLC控制策略。神经网络自适应律采用Lyapunov方法设计,保证了闭环系统所有信号有界。最后采用新方案实现了ASV飞控系统,仿真结果表明整个闭环系统在鲁棒性能方面得到很大提高。

关键词: 飞行控制, 非线性控制, 自适应控制, 神经网络

Abstract:

An adaptive trajectory linearization control strategy and its application to an aerospace vehicle (ASV) are presented. Theoretical analysis illustrates that the current trajectory linearization control method (TLC) lacks enough robustness to the system uncertainties. Firstly, a radial basic function neural network disturbance observer (RDO) is developed. Rigorous proof demonstrates that the RDO has excellent approximation ability to monitor the uncertainties. Then, a novel adaptive TLC scheme is provided by combining the TLC with the RDO output.

Key words: flight , control , system,  , nonlinear , control , system,  , adaptive , control,  , neural , network

中图分类号: